Redeploying the container to 11.1.X.X

Deployment options: Netezza Performance Server for Cloud Pak for Data System

Learn how to redeploy the container to 11.1.X.X on other control plane nodes by using the ips-deploy-remote.sh script.

Procedure

  1. Log in to the e1n1 bare metal node as the root user.
  2. Change directories to the folder in which container.tar.z was extracted to during container upgrade.
    cd /opt/ibm/appliance/storage/ips/ips1_deployment/11.x.x.x
  3. Run ips-deploy-remote:
    ./ips-deploy-remote -n node2 -n node3
    Note: On Cloud Pak for Data System, node1, node2 and node3 are the DNS aliases for the first, second, and third control plane nodes.
    The necessary parameters:
    -n <name>
    The node on which the container needs to be redeployed. For multiple nodes: -n node1 -n node2.
    Optional parameters:
    -m <name>
    Custom domain names that are to be added to the DNS search list in the container. For multiple domains: -m domain1 -m domain2, etc.
    -C <name>
    The container name. The default is: ipshost1).
    -d <path>
    The base directory path. The default is: /opt/ibm/appliance/storage/ips.
    -s <path>
    The directory path to save the Container. The default is: /opt/ibm/appliance/storage/ips/ips1_deployment.
    -I <interface>
    The network interface. The default is: fbond.4079.
    -o <path>
    The Online directory path for Lenovo based systems. The default is: /opt/ibm/appliance/platform/onecli.
    -r <path>
    The srvadmin directory path for Dell based systems. The default is: /opt/dell.
  4. Important: If you are upgrading replication environment as described in Upgrading Netezza Performance Server and Persistent Transient Storage (PTS), do not start Netezza Performance Server and do not proceed with next steps. Ignore this note, if upgrade is not for a replication environment system.

    Start Netezza Performance Server.

    nzstart

Known issues

  1. The docker commit command doesn't complete if the size of the /var/log/tallylog is greater than 1 GB.

    Workaround:

    1. Before you run ips-deploy-remote, check for the tallylog file size:
    docker exec -it ipshost1 bash
    ls -lh /var/log/tallylog
    2. If the size of the tallylog file is bigger than 1 GB, move the tallylog file to a different location with enough space to contain it:
    mv /var/log/tallylog /nz/tallylog_backup
    cat /dev/null > /var/log/tallylog
    ls -lh /var/log/tallylog
    exit
    3. Run the ips-deploy-remote script. After the script completes successfully, move the tallylog_backup file to /var/log/tallylog: 
    docker exec -it ipshost1 bash
    mv /nz/tallylog_backup /var/log/tallylog
    ls -lh /var/log/tallylog